Park and ride is not as significant as you might assume

It’s unsurprising that the closure of station car parks along the Dandenong line for the “skyrail” level crossing removals was highlighted by the media on Wednesday; at this stage it’s the major disruption impact that’s expected.

But – perhaps because of the amount of space it takes up – it’s often incorrectly assumed that Park And Ride accounts for the majority of train users in Melbourne. The stats tell a different story.

Car spaces closing

PTV says the following car spaces will be lost during the project, and I’ve compared that with the total number of train users at each station.

Station Users Car spaces closing % affected
Carnegie 3,140 166 5.29%
Murrumbeena 2,760 292 10.58%
Noble Park 3,790 up to 331 8.73%

This assumes one passenger per car space, which is probably not far off how it works. I’ve also assumed all Noble Park parking is closing, though so far they’ve only said the Mons Avenue car park is closing.

So a total of about 8% at those stations would be directly affected by the car park closures.

Extra spaces opened

To counter the closures, extra spaces will be provided at nearby stations – some also on the Dandenong line, some on the Glen Waverley line.

  • East Malvern 172
  • Holmesglen 170
  • Huntingdale 92
  • Sandown Park 166
  • Clayton 238 (from September)

I make that a net loss of just 49 spaces. (Have I missed something?)

It should be obvious that car park closures have less impact than closing the stations and rail line altogether – as is happening on the Frankston line now.

Car users are the ones most able to adapt their travel patterns to use another station. For instance East Malvern is only 1.5 km from Murrumbeena; only a few minutes drive. That said, if coming from the south side of the line, you’d have to anticipate a delay at the level crossing. And many of the extra spaces created are further out from the inner stations.

In contrast, those who walk or bike to the station are least easily able to switch to another station.

Murrumbeena station

How many Dandenong line passengers use Park and Ride?

Okay so it’s a net 49 spaces removed. But the worst case scenario is that during the project, all spaces at all the stations being rebuilt are removed.

And a lot more people drive to the station than park in the car parks – some of them may be sharing rides, but many would park in nearby streets. Those streets might be affected by parking restrictions during construction.

But even if we assume the worst case scenario, it’s still a minority of the people using these stations.

PTV figures show the following passenger numbers at the stations to be rebuilt.

Station Weekday entries Access by car %car
Carnegie 3140 576 18.34%
Murrumbeena 2760 860 31.16%
Hughesdale 1870 439 23.48%
Clayton 4920 1289 26.20%
Noble Park 3790 1174 30.98%
Total 16480 4338 26.32%

So perhaps 4338 people (26%) affected by car park closures, if all the car parks close, and if nobody can park anywhere in surrounding streets within walking distance… unlikely.

Looking at the entire line

In contrast, how many would be affected if the line closed completely for weeks or months at a time? Not just those at the closed stations, but also all of those coming in from further out, who would face a train/bus/train trip.

As the figures below show, assuming they all come through the sections to be closed, it’s about 64,000 people per weekday — fifteen times the number affected by the car park closures, and more than double the number affected by the Frankston line closure.

The figures also also show that while a minority drive to the stations slated for rebuilding, the amount of Park And Ride increases as you get further out.

Weekday entries Car %car
Carnegie 3,140 576 18.34%
Murrumbeena 2,760 860 31.16%
Hughesdale 1,870 439 23.48%
Oakleigh 6,780 1,701 25.09%
Huntingdale 6,260 2,165 34.58%
Clayton 4,920 1,289 26.20%
Westall 1,980 724 36.57%
Springvale 4,690 1,287 27.44%
Sandown Park 1,900 1,383 72.79%
Noble Park 3,790 1,174 30.98%
Yarraman 1,010 466 46.14%
Dandenong 8,300 2,222 26.77%
Lynbrook 1,270 491 38.66%
Merinda Park 1,160 847 73.02%
Cranbourne 2,130 1,036 48.64%
Hallam 2,310 1,439 62.29%
Narre Warren 2,830 1,612 56.96%
Berwick 3,190 1,603 50.25%
Beaconsfield 840 545 64.88%
Officer 70 47 67.14%
Cardinia Road 900 462 51.33%
Pakenham 1,910 1,193 62.46%
Total/Average 64,010 23,561 36.81%

So basically the further out you go, the higher the proportion of park and ride users. Which probably reflects the overall walkability of those suburbs, and of course the number of residences within walking distance of the stations.

It is a lot of people, but it’s still a minority across the entire rail corridor.

Bus at Murrumbeena

More park and ride? Or better feeders?

Of course station car parks are expensive to build (tens of thousands of dollars per space) and not a great use of land – the space they take up certainly doesn’t enhance walkability.

This article about a US survey of park and ride notes numerous problems (though under-utilisation isn’t really an issue in Melbourne).

There is huge scope to improve local feeder buses. This goes doubly for the railway stations having their car parks temporarily lost – rather than spending up big extending car parks elsewhere, they might have done well to fund additional feeder bus services, particularly in peak hour. (The Murrumbeena bus used to run twice as often in peak hour.)

The removal of the crossings provides a great opportunity to upgrade buses. With far fewer delays at crossings once they are removed, bus punctuality and efficiency will improve a lot. PTV and the government should take the opportunity to give more people a way to use public transport without having to own a car.

  • Update: Quite possibly affecting more people at Murrumbeena than the car park closure is the closure of the pedestrian footbridge, which is used by many to cross the tracks (either in the morning or the evening) to avoid the level crossing. This morning we’ve learnt: Murrumbeena ped bridge will close mid-August and will be demolished shortly after.
If you enjoyed this post, please consider leaving a comment. You can subscribe via feed reader RSS, or subscribe by email. You can also Follow me on Twitter, or Like the blog on Facebook.

17 Replies to “Park and ride is not as significant as you might assume”

  1. A good research piece as usual, but a couple of personal observations:
    * As a life-long commuter into the city by train, I would never live further than walking distance from a station. I would never want to wait for a bus that may or may not turn up, then wait for a train which I probably just missed because the bus driver caught three red lights instead of two. I would drive all the way into the city instead. Luckily, I can afford to live near a station.
    * Living in a household of adult women, none would want to walk home from the station in the dark or wait at a dodgy bus stop for a bus that may or may not turn up. And then walk home from the bus stop. They would phone ahead and get someone to pick them up (they would have driven to the station except the station car park would already be full).
    So I think there are human preferences that prefer cars in many situations and have had bad experiences with bus-train connections

  2. I noticed the electronic counter for car park spaces remaining at Gardiner station hasn’t been working the last several times I’ve walked past. Maybe they’re embarrassed at how many free spaces aren’t being used? Possibly the entrance being relatively isolated from Burke Rd isn’t helping, or commuters just got used to the alternatives on offer during the disruption.

    I’m surprised the car percentage for Merinda Park isn’t higher. It’s a disgrace that there’s no footpath towards Thompson Rd. Westbound / Evans Rd with all the new housing estates being built.

  3. Tom Minear authored a “story” in the Herald Sun today about the impact of car park closures on traders. It suggests that closing train station carparks is going to hurt traders and force closures. Which is ridiculous, because those carparks usually fill up with commuters early in the day and are not available for shoppers. Futhermore, train commuters can’t park near shops due to restrictive time limits, and shoppers are not legally permitted to use train station carparks.
    http://www.heraldsun.com.au/news/victoria/melbourne-sky-rail-jobs-to-be-lost-small-businesses-crumble-during-construction-chaos/news-story/0247ae254edad7fa59ca96664f5180c5
    He fails to acknowledge that the surrounding roads and the train stations themselves will remain open – moreso than they would under a rail-lowering scenario – in a deliberate effort to reduce disruption. With roads and stations open, people can still come and go through the neighbourhood, and it will mean business *mostly* as usual for traders. If he wanted a comparison scenario, where traders were actually impacted by lengthy road closures, station closures and carpark closures, Tom would only have to visit Ormond, McKinnon and Bentleigh, where the rail-under-road works are reaching a crescendo after 10 gruelling months of disruption.

  4. @Roger, I’m going to look more closely at the stats with regard to bus/tram feeders, but you’re no different to anybody else… an infrequent (and yes, quite possibly unreliable) bus just isn’t a viable connection for most people. It’s a different story for your fellow commuters at Elsternwick… the tram is about as frequent as the train most of the day, such that 18.6% of train passengers reach the station by tram.

    @nick, that’s interesting. I wouldn’t be surprised if usage builds up gradually again. Perhaps they should be promoting Gardiner to Dandenong line park and riders?

    @Michael, there is impact on some traders however. Apparently some shops on Victrack land are to be demolished as part of the project.

  5. interesting – I’d noticed the extension works being done to Holmesglen, East Malvern and Darling station car parks but hadn’t twigged that they were to cater for works on another line … it’s too early to tell either way, but I wonder whether they are intending on leaving them as gravel/dirt surfaced rather than fully sealed as this (presumably) would cost less for a temporary spike in commuter numbers coming from other stations …

  6. @daniel @nick

    Gardiner station carpark always seemed to be used by people coming from the Monash freeway, I wouldn’t be surprised if someone put a VMS on the freeway advertising the fact that the carpark had reopened and a week later it being full again. Source: I work very close to the station and commute using it 50-60% of the time

  7. We did number plate recognition surveys at Gardiner Station pre works and found that 50% of cars in the carpark had travelled from further out on the Monash from Glen Waverley, Rowville and Narre Warren – remember it used to be a Zone 1 station so was a cheaper alternative than travelling from further out and had great access from the Monash

    I think the project works have changed some habits. It also has a good tram connection and is in a walkable catchment.

  8. Problem is often I find is park ride is the local community and the people who live near the station. Like if your local resident and there isn’t enough park ride spots the excess cars will spill over into local streets making a life a nightmare for locals and you the resident.

    Though now people who live near railways stations have been really getting into renting out their car spot/garage. Which is smart.

    I’ve found as someone who lives in near a busy Rail station that doesn’t have a park and ride it can be impossible to find a park. I sometimes don’t use the car in fear of loosing my spot.

    Problem with feeder buses is they’re slow since they’re in frequent and aren’t direct. If they are frequent and direct they often run empty at heavy tax payers expense.

  9. Feeder buses really are the only way to bring large amounts of people onto the railway network. Parking might be somewhat viable at outer suburban locations, but at the end of the day, with the projected passenger numbers, having parking for a lot of them won’t be viable.

    At the same time though, the piecemeal approach to making feeder buses work isn’t really effective. As many people both here, and in general point out – feeder buses are too infrequent and slow to be useful, and tend to run empty. Even the isolated upgrades taking place at various locations around the network aren’t enough. Making feeder buses work means that real thought has to be put into it, from the location and amenity of interchanges, to having strategies for servicing new estates. It also means that it isn’t enough to make upgrades to a few suburbs every couple of years. If I can take a bus from my place, but the buses aren’t useful at my destination, I won’t be able to make use of the feeder bus from my station, no matter how useful it is.

    There have been some good improvements, and hopefully some good results from them – but it will take more than a couple of small, isolated and scattered upgrades to make a real difference.

    Plus, we’re still waiting for the Road-based PT Plan, 3 years after the rail one emerged…Christ.

  10. Ilham, the council round here deals with this by making one side of the road a 2 hour parking limit except for residents. There is consequently usually plenty of parking on the road for residents. Some residents then, of course, lend or sell their parking permits to commuters.

    Still doesn’t stop some residents complaining because they can’t necessarily park in ‘their’ car park outside their house.

  11. The carpark and train are usually almost full by the time I catch a train in the mornings (a bit after 7.30am at Carnegie) so I don’t know if it is people parking to use the shops or driving in. The few times I have gone in much earlier it is mostly tradies on the train heading into construction jobs or shift workers.

  12. An insightful article Daniel. Dr John Stone and I wrote a paper on some of these issues for last year’s ATRF conference. It investigated the relationship between feeder bus service levels and rates of transfer to rail on the Dandenong line section being updated for skyrail. Unsurprisingly, the evidenced showed a positive relationship between frequent bus services, low connection wait times and higher rates of transfer.

  13. I agree on two points with the feederbuses. Firstly, we need to focus more on them, and we are a long way from a decent network of the same.

    I would like to see the stats on stations which have SmartBus routes connecting with them, and are their patronage same or similar to trams.

    RE:Roger, with adult women; Perhaps we need such bus stops to have ‘SmartBus style’ PIDS, so at least people will know how long the next bus is. If it says, next bus is in 20 minutes, you may as well head off and stand elsewhere for 10 to 15 of that.

  14. Buses are a poor answer.
    If you postulate that most drive-and-ride train users live between 1 and 6 km from the station [ people less than 1 km are most likely to walk ), then how many bus routes would you need, to have a bus route accessible to people living 5 or 6 km away in any direction from the station ?

    The answer is, too many !

  15. The majority of people taking trains would live within about 2-3km of the station – any further and a different station and line may be closer, or using a different mode altogether may be more effective.

    Of course, at 5-6km distances from stations, it is quite difficult to have effective feeder services, but that situation is pretty rare anyway, especially 5-6km in every direction – it is much more likely to occur in a general direction, and only at the ends of railway lines anyway.

    With the bulk of passengers living in a reasonable distance (1-4km away) that may have to park and ride, buses offer a reasonable solution. You can never build enough parking – we could spend the entire budget for the Metro tunnel on parking , and soon it wouldn’t be enough, and the improvement in service and mobility would be atrocious for the cost.

  16. @enno – how does that logic work? If people will walk 3/4 of the distance to a bus stop that drops them right at the station, a single bus route will draw from a 1.5km wide corridor along its path to the station. That’d be about the average distance between inner stations, so 1 route per station… if the route continued would almost certainly intersect with several other stations on the radial network.
    Call it 2 routes per station further out.
    People will walk 750m for good service that connects to the train. Problem is, we don’t often offer that.

  17. Thanks for the article Daniel. I agree that the emphasis has been on people who drive and park without much consideration of others, and sometimes to their detriment. For example, near Clayton station they closed the shared pedestrian and cyclist path on the north side of the train line with no warning or information. When I contacted them, I was told it was for the duration of the works (2 years) and the detour is to just use the road (for people who ride). I have no problems walking or riding home from the station in the dark, so long as there’s a place for me to do so.
    In contrast, the commuter parking on the south side fills up once a year (for the football grand final at the oval opposite) and there appears to have been no consideration of closing that.

Leave a Reply

Your email address will not be published.